Core Definition, Boundaries & Business Drivers
What Data Governance is
Chapter 3 defines Data Governance as the exercise of authority and control over the management of data assets. The definition includes planning, implementation, monitoring and enforcement.
The word authority matters: Governance decides how decisions about data are made, who is accountable, what rules apply, how conformance is monitored and how unresolved conflicts escalate.
Governance vs Data Management
| Data Governance | Data Management |
|---|---|
| Ensures/directs/oversees how data is managed. | Performs the operational/specialized work used to manage data. |
| Decision rights, policy, standards, monitoring, enforcement, escalation. | Modeling, storage, integration, quality, Metadata, operations and other Knowledge Area execution. |
Figure 15 lesson: Governance is not a super-team that personally performs every technical or operational task. It creates authority and oversight so the wider Data Management system operates coherently.
Deciding clue
- “Who decides, approves, owns, sets policy, monitors or escalates?” → Data Governance.
- “Who builds, operates, remediates, integrates, models or manages technically?” → Data Management execution.
Data Governance vs IT Governance
Data Governance focuses on the data asset — how data is defined, protected, used, shared, valued and managed.
IT Governance focuses on IT investments, application portfolios, project portfolios and technical architecture.
A data problem may use technology without becoming an IT-governance problem. Ask what object is being governed.
The eight common DG scope areas
Chapter 3's introductory scope includes governance over areas such as:
- Strategy.
- Policy.
- Standards.
- Oversight.
- Compliance.
- Issue management.
- Data Management projects.
- Data asset valuation.
These are not eight isolated departments. They are areas in which governance supplies authority/control.
Business driver: governance is not an end in itself
Chapter 3 groups the business case into two broad families:
Reduce risk
Examples include financial/reputational risk, security/privacy, regulatory/legal exposure and mishandling of data.
Improve processes
Examples include regulatory response, Data Quality, Metadata/Business Glossary, development efficiency and vendor/data-acquisition management.
The best governance program starts from measurable business pain/opportunity. “Governance is best practice” is not a strong business case.
Why technical initiatives expose governance need
A customer MDM initiative may require decisions about: - authoritative source; - merge/survivorship rules; - definition of Customer; - Data Quality thresholds; - ownership/stewardship; - issue and exception paths.
Matching software cannot legitimately invent these enterprise decision rights. Technology implementation can reveal the need for governance.
Data-centric organization
A company is not data-centric merely because its databases are modern. A data-centric organization treats data as a corporate asset distinct from the systems/infrastructure that host it, aligns data strategy to business strategy, pursues quality and continuously improves Data Management practices.
Stop and check
A DGC starts approving every database patch and running remediation scripts. What went wrong?
The governance body collapsed oversight into execution. It should define/approve decision rules and handle appropriate escalations; technical Data Management teams perform controlled operational changes.
Source anchors: Chapter 3 pp. 69–75; Figures 14–15.